Are humanoid robots the most effective robots we can build?

The etymology of ‘engineer’ is ‘ingenuity,’ so why are we so focused on humanoid robots?

We’re already living in a post-humanoid era, with billions being poured into embodied, humanoid robotics right now. According to a 2024 Goldman Sachs report, the market for humanoid robots is projected to reach $38 billion by 2035 and Morgan Stanley anticipates $5 trillion by 2050. From the World Humanoid Robot Games, to Tesla’s Optimus folding laundry and Amazon’s tests of robots that hand-deliver packages, 2025 has seen an unprecedented acceleration in humanoid robotics demos; exciting and impressive, but also fueled by a lot of ‘hype.’ While these headlines are exciting, what if our obsession with humanoid robots is slowing down real progress?

It’s a case of Dr. Ian Malcolm’s famous warning come to life: “Your scientists were so preoccupied with whether or not they could, they didn’t stop to think if they should.” The same applies to robotics today. Why are we so intently focused on mimicking the human form?

Here’s the problem: the human form is a poor blueprint for automation. Humanoids aren’t the most effective or efficient robots we can build.

• Energy inefficiency: Battery-operated mechanical devices at this scale and size would consume massive amounts of energy (high double-digit percentages) just to maintain upright movement and humanoid functions.
• Human dexterity is incredibly difficult to replicate. Engineers still struggle to replicate a grasping human hand, or the hand-eye coordination of a human brain.
• Cost and complexity: These machines are incredibly expensive and fragile. They’re built for demo reels, not scalable, real-world deployment.

Even if we solved for the challenges above, why should we limit ourselves, creatively, to the human form factor? Why two arms? Why not four arms?! Why two legs instead of wheels? Our world is built for human bodies, but that doesn’t mean our robots need to take our exact same form.

Unpacking the humanoid hype

Yes, our human world is built and shaped for human bodies. There is a popular fantasy of $20,000 humanoid robots being mass-produced; suddenly cheaper, owned by the masses, and capable of solving all of our biggest operational problems and (possibly more popular) our household chores.

One part of the fascination with humanoid robots reflects our reality: we don’t have enough people interested in factory work. As of March 2025, there were half a million unfilled jobs in manufacturing in the U.S. alone, and the North American manufacturing workforce has declined for decades. The logic follows, if we can’t hire more people, we must build more people to fill these gaps.

I also think, in part, our fascination with the humanoid form is ego-driven. As we mechanize our own lives with LLMs and chatbots, we are approaching the uncanny valley. We have entered into that twilight zone between what we know is living and what we know is mechanical – that middle ground where the ‘real’ and ‘unreal’ cross over each other and blend together.

Two decades of overhyped automation has proven that robots can’t replace humans at scale. Robots aren’t standalone solutions, they’re components in a larger, interconnected production system. They require integration with workflows, data infrastructure, human supervision, and continuous maintenance. And while one robot might be manageable, scaling to a fleet across lines, shifts, and changeovers remains a complex, unsolved challenge. Robots are tools, not strategies. The real value comes from how they fit into the system around them.

The etymology of ‘engineer’ is ‘ingenuity’

Nassim Nicholas Taleb says, “Missing a train is only painful if you run after it.” In other words, if we’re losing the humanoid robotics race to China, maybe we’re chasing the wrong race. If the Western world is already lagging behind China in the global humanoid robotics race, why don’t we modify the ‘rules’ of the game to put us back on track?

I think our advantage has always been, and will continue to be: ingenuity. Take a moment to consider the word “engineer”: not as a job title, but in its etymology. It comes from the Latin ingenium—meaning ‘innate talent,’‘clever invention,’ and ‘ingenuity’. At its core, engineering isn’t meant to be rigid execution—it’s about applying human creativity to solve our problems. It’s our human ability to think innovatively, quickly problem-solve, and shift direction without losing momentum.

Innovation naturally emerges when people tackle challenging problems in interesting ways. Google, for example, was founded as a digital replacement for the Dewey Decimal system, and now it delivers global access to information in milliseconds. Most innovations aren’t instant breakthroughs; they don’t involve big leaps, but thousands of incremental steps.

From an engineering standpoint, humanoid robots don’t yet make practical sense for most businesses. If we’re chasing after sci-fi prototypes and geopolitical milestones, we might be missing that layer of human ingenuity; what could we create now that would serve the same function, but faster, more utilitarian, and energy efficient?

Where Robots are Already Winning

When I look at the immediate impact of automation and robotics, I’m most impressed by the robotics companies that are focused on functionality and versatility. Robots don’t need to look human to deliver real value.

Many of the most successful and widely adopted robotics programs in manufacturing aren’t humanoid at all; they’re purpose-built for specific tasks, not designed to resemble humans.

• Universal Robots’ UR series robotic ‘arms’ are designed to collaborate safely with human workers in shared workspaces. These robots handle precision tasks like screwdriving, gluing, and component assembly, all without the need for fencing or isolation. Their lightweight frames and intuitive programming interfaces make them ideal for high-mix, low-volume production environments.
• For quality inspection purposes, Cognex or Zebra Technologies leverage IIoT computer vision, and machine learning to scan components for micro-fractures, misalignments, surface defects or incorrect labeling in real-time.
• DMG MORI’s AMRs (Autonomous Mobile Robots) driverless transport systems integrate with existing production systems to help transport workpieces, pallets, and tools, helping operators stay focused on higher-value tasks.

These systems are succeeding not because they look like humans, but because they solve specific problems with precision and efficiency.

If manufacturing jobs are simply too difficult to fully automate with humanoid robots, today’s opportunity for innovation is task automation. If we can build machines to perform tasks, not entire jobs, this can help fill the skills shortage gaps, while also improving conditions and career progression for our frontline operators and engineers.

The future of robotics isn’t humanoid, it’s human-driven. The most transformative solutions will come from those who focus less on copying us, and more on complementing us.

If it’s an ‘arms’ race, it’s up to us to decide what type of ‘arms’ make the most functional sense.